Knowledge Management in German Industry

Transcription

1 Knowledge Management in German Industry Study in the Framework of an OECD Initiative of the Centre for Educational Research and Innovation (CERI) on behalf of the Donors Association for the Advancement of Science in Germany (Stifterverband für die Deutsche Wissenschaft) with the support of the Federal Ministry for Education and Research (Bundesministerium für Bildung und Forschung/BMBF) FINAL REPORT Dr. Jakob Edler Fraunhofer Institute for Systems and Innovation Research (ISI) Karlsruhe January 2003

2 Contact: Dr. Jakob Edler Fraunhofer Institute for Systems and Innovation Research (ISI) Breslauer Str. 48 D Karlsruhe Tel Fax Rainer Frietsch from Fraunhofer ISI has contributed to this study, especially with indispensable support as for the statistical analysis.

3 Content List of Figures...III List of Tables...IV 1. Introduction: Context and Scope The growing meaning of knowledge and the new needs for knowledge management Knowledge gaps in industrial knowledge management in Germany Objectives of this study The Sample Relevance of a Broad KM Concept: A Non-response Analysis The Employment of KM Practices in German Industry Overall pattern Comparative analysis of the diffusion of KM practices The recent and future dynamics of KM practices The Driving Forces of KM: Motivation Patterns in German Industry Overall pattern Comparative analysis of KM motivation Effects of KM Overall effects Comparative analysis of KM effects The Institutionalisation of KM and its Meaning for the Use of KM Different levels of dedication towards KM The effects of institutional commitment KM and its Role within Innovation Management...39

4 8.1 Employment of Knowledge Management: Innovators vs. Non-Innovators The Capture of Knowledge for the Innovation Process Introduction The growing meaning of knowledge absorption Drivers to utilising external technological knowledge Usage and meaning of sources for external technological knowledge Obstacles to use external technological knowledge Diffusion and meaning of knowledge about external technological knowledge Practices to manage the usage of external technological knowledge Characteristics of users of external technological knowledge The effects of practices to manage external technological knowledge Summary and Conclusions KM in general KM to capture external knowledge for the innovation process Future challenges for the analysis of KM...79 Literature...80 Statistical Annex...83

5 List of Figures Figure 1: Size distribution of the sample (%): total and sectors...12 Figure 2: Average number of KM practices used - size...16 Figure 3: Figure 4: Figure 5: Figure 6: Figure 7: Figure 8: Average number of KM practices used sectors...17 Usage of KM measures mean value for KM categories (indices) sector...18 Importance of Cluster of KM Motives - size...27 Importance of cluster of KM Motives - sectors...28 Cluster of KM Effects - size...31 Cluster of KM Effects - sectors...32 Figure 9: Institutionalisation of KM - size...34 Figure 10: Institutionalisation of KM - sector...35 Figure 11: Figure 12: Figure 13 Figure 14: Figure 15: Figure 16: Figure 17: Figure 18: Figure 19: Figure 20: Figure 21: The meaning of KM budget for KM effects...36 The meaning of centralised KM organisation for KM effects...37 Usage of external technological knowledge...42 Importance of sources for technological and market knowledge - size...43 Importance of sources for technological and market knowledge - sector...44 Importance of reasons to use external technological knowledge - size...46 Importance of reasons to use external technological knowledge - sectors...47 Usage of practices to acquire external technological knowledge...48 The meaning of sources for external technological knowledge...49 The meaning of obstacles to use external technological knowledge...56 Knowledge about external knowledge whole sample...58

6 Figure 23: Knowledge about external knowledge services...59 Figure 24: Figure 25: Figure 26: Figure 27: Figure 28 Companies employing practices to manage the usage of external technological knowledge...61 The non-usage of specific practices to manage externa l technological knowledge...63 The importance of obstacles frequency of using external technological knowledge...66 Knowledge about different external sources...68 Impact of practices to define long-term demand on selected KM effects...70 List of Tables Table 1: Company sample and response rate sectoral distribution...11 Table 2: Non-response analysis...13 Table 3: Percentage of companies using selected KM practices total sample...15 Table 4: Table 5: Table 6: Motivations to use KM, whole sample...23 Definition of factors: motivation for KM...25 Effects of KM whole sample...30 Table 7: Growing meaning of external technological knowledge - size...45 Table 8: Growing meaning of external technological knowledge - sectors...45 Table 9: Table 10: Table 11: Table 12: Meaning of different instruments to source external technological knowledge...54 Practices used to monitor the external relevant supply of technological knowledge...61 The meaning of using external technological knowledge frequently...65 Innovation effects of practices to manage external technological knowledge...69

7 1. Introduction: Context and Scope The notion of a knowledge-based economy points towards what has become the most important asset of OECD economies: knowledge. For some years now, international organisations such as the OECD and the EU have put knowledge and its management at the centre of many activities and analyses.1 In order to systemise the knowledge on knowledge management (KM) within the OECD, the Centre for Ed u- cational Research and Innovation (CERI) of the OECD started an international study initiative in the year The motivation for this initiative was based on the reflection that up till now too few internationally and intersectorally comparable data are available on how enterprises generate and productively implement knowledge and how they keep it permanently accessible. In addition, the studies which meanwhile exist use very diverging definitions and concepts concerning knowledge management and are for the main part based on case studies. The long-term goal of this OECD initiative will be to create for Knowledge Management (KM) similarly as for Innovation (the Oslo Manual) and for R&D (the Frascati Manual) a uniform, binding manual and thus to structure the international discussion on this subject. This study is the German contribution to this international OECD undertaking, based on a survey in seven industrial sectors, including services. Therefore, its concept is to a large part the result of a moderation within the international study group of the OECD. The questionnaire used for the survey contains a core that was binding for all country studies. As the study is exploratory, the understanding of KM is broad and covers the usage of 19 rather diverse KM practices, as well as the motives for and effects and institutionalisation of KM. In addition to this uniform core study, the study on hand goes beyond the OECD core and has integrated a series of questions that relate to a specific dimension of KM, i.e. the management of interfaces to capture knowledge from outside the company. To be sure, there are many aspects of KM that would deserve a deeper analysis. The motivation for the selection of this special focus stems from the fact that while the importance of external knowledge for the performance of industry as regards innovativeness and market success is well known meanwhile, not much systematic knowledge exists on how this sourcing across interfaces is managed. But if external knowledge, especially scientific and technological knowledge, has an impact on the performance, and if we share the premise that it becomes even more 1 Most recently, the OECD organised a Global Forum on the Knowledge Economy (OECD 2002). 2 These empirical studies on knowledge management in enterprises are incorporated in wider activities of the OECD on the subject of knowledge management. They comprise among others high-level seminars, regular scientific activities of a high level study group on knowledge management and a systematic review of national education systems from the perspective of knowledge management.

8 2 important in the future, the management of it is a core pillar of knowledge management as well as of innovation management. The following introducing chapters explain the meaning and concept of KM (1.1), present the objectives of the study (1.2) and point towards the related knowledge gaps on KM in German industry (1.3) The growing meaning of knowledge and the new needs for knowledge management To claim that the meaning of knowledge and its management has grown meanwhile is a commonplace. But in order to understand the importance of managing knowledge and to tailor appropriate analytical concepts, it is important to understand the major drivers for this enhanced meaning of knowledge. These drivers are of course numerous, and the relative weight of each of these drivers differs with the context of companies as well as with the concepts of knowledge and KM in mind. Without any implicit ranking, the major drivers for the enhanced meaning both of knowledge and its management that can be found in the literature are as follows:3 We observe a growing stock and flow of knowledge that needs to be taken into account by decision-makers. Knowledge has become a commodity that is traded and transferred in numerous new ways. This not only necessitates the management of knowledge flows, but leads to the creation of new knowledge through re-combination (e.g. Cowan et al. 2000). There is not only greater demand to trade knowledge as a commodity, but this can be done with decreasing costs. Information and Communication Technologies (ICT) lower costs of knowledge storing, seeking, employing and transferring. The acceleration of the production of knowledge and shortening of product life cycles and related innovation cycles increases the pressure on the processing of knowledge. Due to increasing fluctuation in industry and the increasing mobility of knowledge workers, there is a growing danger of costly loss of knowledge and buildup of new knowledge in companies. It is therefore becoming more important to document knowledge and to make it available to newcomers. 3 The literature on the meaning of knowledge has become abundant. The major drivers enumerated here are, among others, based on Cowan et al. 2000; OECD 1999, 2000, 2002, Worldbank 2002.

9 3 The borders of organisations are changing more rapidly, mainly due to increased merger and acquisition activities. This leads to ever new demands to integrate existing knowledge stocks and to create new, integrated knowledge flows. The two preceding tendencies especially put pressure on the Human Resource Management (HRM) of companies, as the workforce must be enabled to manage, update and store the knowledge flow relevant to their work. Above all, individual knowledge needs to be transferred into organisational competencies in a sustainable way, i.e. not only the workers, but the organisation must take up and process the knowledge available. The changing role of knowledge has, above all, severe implications for innovation and innovation management: Almost ten years ago, analysts observed a changing mode in the production of (technological) knowledge ( mode 2 4, Gibbons et al. 1994), characterised by heterogeneous interaction and hybrid interaction forms. This demands among other things openness of organisational knowledge production systems to the environment. Above all, appropriate organisational interfaces are called for that enable the members of an organisation to interact internally, as well as with outsiders. These interactions include especially the interaction between companies and research sources such as public research institutes. Generic technologies, most obviously ICT or nanotechnology, have led to a fusion of technological disciplines and knowledge areas, thus triggering new demands to integrate knowledge across a wide variety of knowledge areas and disciplines that used to be separate. Innovation must increasingly be understood as a re-combination of existing knowledge or as a consequence to utilise and exploit existing knowledge sources effectively (among others Foray/Gault 2002; Foray 2002). The transfer of knowledge as a commodity and the re-combination of knowledge leads to creation of new knowledge to foster innovation. This constructive process demands a management of the acquisition of knowledge, both from outside and from within companies. In addition, the economies of speed do not allow companies to generate in-house what already exists elsewhere. Moreover, as the complexity of technologies and the underlying scientific knowledge has grown, companies are less and less capable of generating all knowledge needed for the innovation process in-house, e.g. within their R&D departments and laboratories (Barabaschi 1993). 4 Mode 1 production of knowledge refers to the traditional, more linear model of university-based, pure knowledge creation that diffuses through the system downstream and is, at the latest stage, applied to the market. In mode 2, by contrast, knowledge is produced largely as a result of demands expressed from the application side, it is transdisciplinary, entrepreneurial and embedded in networks (Gibbons et al. 194, see also OECD 1999, p. 61).

10 4 It pays off to internalise external knowledge, and it is especially beneficial to use scientific knowledge. The consequence of the increased interlinkages between the innovative activities of companies and external knowledge providers have already been analysed on an aggregate level. Janz. et al. (2000) have analysed the meaning of four different sources for the innovation process in depth. They have impressively shown that the usage of external sources makes a difference as for the innovation and thus for the economic performance of companies. In the manufacturing sector, users of scientific knowledge have the highest market share with innovative products and even more striking expect employment and sales to grow by far the most compared to users of other sources. At the same time, the share of companies using science as a source is lowest as compared to other sources. Similar results are shown for the service sector, albeit to a lesser extent (Blind et. al. 2000, especially p. 49). Given these obvious effects of using external knowledge, the management of the processes to monitor and internalise external knowledge are crucial. The interface between companies and especially between companies and the science system need to be a major focus both of analysts and corporate strategists. A major objective must be the increase and optimisation of the usage of external knowledge for the innovation process. What are the consequences of all this? In light of all these challenges and opportunities at different levels, a comprehensive knowledge management is necessary which must exceed the classical instruments like the maintenance of in-company databases, a well-functioning internal reporting system, and periodical further education/training measures. The aim must be to establish a self-renewing knowledge cycle that enables organisational learning and the exchange with the environment. It is increasingly necessary thereby to process the knowledge distributed throughout the whole enterprise and make it accessible to all employees, to minimise the loss of company knowledge through the increased fluctuation of personnel. Furthermore, the knowledge cycles in enterprises must constantly be fed by knowledge and ideas from outside the firm, just because of this broadening of the knowledge base for innovations. The generation of the total relevant technological knowledge can increasingly no longer be performed in a single enterprise on its own. The practices to absorb external knowledge which have already been in existence for years, through Public Private Partnerships, cooperations, contract research, publications etc. must be extended, systematised and made more effective, against this background (interface management).5 This applies particularly to SMEs, which are confronted with similar technological challenges as the large enterprises but which can only enlarge their knowledge base to a certain extent autonomously, due to more limited capacity. 5 For an early recognition see for example Barabaschi (1993): Managing the Growth of Technological Information; in: Rosenberg et al. (Ed.), Wealth of Nations; Stanford: Stanford University Press; pp

11 5 The obvious micro-economic necessity of an effective inter-face and knowledge management is accompanied by diverse macro-economic benefits. Potentials for cost savings lie for example in avoiding duplicated research, or in the multiple utilization of once financed knowledge gain, the cross-pollination of public research and industry, or in the minimisation of so-called sunk costs through knowledge loss as a result of the flexibilisation of jobs. Even although numerous enterprises already have impressive knowledge management systems in place6, it appears that the necessary awareness for comprehensively understood knowledge management can still be improved across the industrial spectrum as a whole (see among others Schmoch et al. 2000, Brockhoff 1999). To sum up, the capacity of firms, industries and countries to develop and manage knowledge assets is a major determinant of competitiveness and economic growth (OECD 2002, p. 2). The consensus on that analysis both within the academic world and with policy-makers is overwhelming.7 The empirical analysis of the meaning of knowledge and its management requires a definition of knowledge management that is able to capture its width and complexity, including the management of knowledge interfaces. Therefore, this study uses the broad definition of KM developed within the OECD core group according to which knowledge management (KM) involves any activity related to the capture, use and sharing of knowledge by the organisation. Above all, this broad definition means that KM for a variety of reasons not only relates to the management of codified information with the help of IT processes. ICT and ICT-related communication is important, but should not be misunderstood as the major or even sole dimension of KM. In addition, the importance of human resources as the carrier and transmitter of knowledge is growing, both as related to KM practices (training for KM), as well as other functional knowledge that needs to be shared with others. Finally, there are indications that, in order to cope with the growing dynamics and complexity of knowledge development, companies increasingly have to rely on knowledge that cannot for reasons explained above be produced within the company itself. In fact, in some cases the acquisition of external knowledge has been defined as crucial for the persistence of an efficient evolution and innovative capacity of companies.8 It is crucial to learn if the companies 6 Cf. e.g. the case studies in Mertens et al. (2001), Knowledge Management. Best Practices in Europe; Heidelberg: Springer-Verlag or Bach et al. (2000): Business Knowledge Management in der Praxis; Heidelberg: Springer. 7 To mention only a couple of key studies and analyses: OECD (1999), Leonard-Barton (1995), Prusak (1997), Davenport/Laurence (1998), de la Motte, J./Foray, D. (2001) Willke (1998), Den Hertog/Huizenga (2000), Calvo/Sancho Munoz 2002). 8 An early recognition of this was made by Barabaschi (1993), a former manager of a large Italian company in the electronics sector.

12 6 are systematically dedicated to KM, i.e. if they have formulated KM strategies, if they have an appropriate value system etc.9 Consequently, the approach of this study encompassed an ensemble of practices ranging from IT solutions for internal storage and communication of data to training and mentoring, from KM strategy plans to practices of knowledge acquisition, and had a special focus on external knowledge interfaces. 1.2 Knowledge gaps in industrial knowledge management in Germany For German industry as for the industry of many other countries a survey applying a broad concept of KM and covering a wide range of sectors was overdue for several reasons. First of all, almost all empirical work done on KM practices in Germany is based on case studies (e.g. Willke 1998).10 As many of these case studies are limited to one key aspect of KM, i.e. ICT-based approaches (Bach et al. 1999, Bach et al. 2000), even the aggregate of case studies cannot provide a general picture of KM in Germany. A second open question regarding our understanding of KM in German industry is, does KM mean different things in different sectors and for different company sizes? There is only one survey that besides a couple of European firms, mainly includes German companies (Heisig/Vorbeck 2001). This very valuable work is limited to some 140 German companies and therefore does not differentiate the answers according to different sectors and sizes. Only from the response rate did the authors find indications that in very general terms KM is apparently used more broadly in certain industries - such as chemistry and pharmaceuticals, consulting, automobiles, ICT and mechanical engineering (Heisig/Vorbeck 2000, p. 121). In addition, although it has been shown that the usage of KM practices correlates with size, i.e. KM is used more widely in larger companies,11 a broad empirical analysis for Germany is still lacking. Thirdly, the existing studies and this is true not only for German companies are focused mainly on the internal KM processes and somehow neglect the interface between internal and external knowledge sources and knowledge processing. However, one central premise of this study is that due to a number of reasons growing complexity, interdisciplinarity, economies of speed, inter-organisational co- 9 In this section, the German study expanded the OECD core questionnaire and added the questions on top management responsibility, respectively incentives (see table 2). 10 Some German cases can also be found in Mertens et al See for example Prusack 2001.

13 7 operation etc. internal knowledge generation is under pressure and must increasingly integrate external knowledge quickly and smoothly. Fourth, the added value of this study on Germany, and especially the dimension of KM and innovation, results from the fact that for the first time, the practices to absorb knowledge for innovation purposes are not only analysed in more detail, but also put into the context of knowledge management in companies in general. In their analysis of innovation sources Janz et. al (Manufacturing) and Blind et al (Services) not only conclude that external sourcing especially but not exclusively from science has a high positive leverage as for innovative capacities. They, moreover, conclude in demanding better absorptive capacities within companies in order to exploit the valuable source science. To analyse the state of the art as for the management of this important absorption will be a central focus of this study. By doing so, the study will fifth counterbalance existing analysis of this transfer especially from the research system to the companies. Existing studies and especially programmatic papers have reviewed the transfer of knowledge and techno l- ogy between the different spheres of the German innovation system by almost exclusively analysing the activities of public research institutions. In contrast to the German private companies, public research institutions as a major source of external knowledge for companies, have been analysed again and again in order to improve their ability to transfer knowledge (recently see Schmoch et al. (2000); Edler, Schmoch (2001)). More relevance and more openness is demanded (e.g. Nicolay/Wimmers 2000; Wissenschaftsrat 2000). Especially business federation put the burden of more effective transfer of knowledge on the research organisations (e.g. ZVEI 1999, BDI/DIHT 2001). However, Reinhard (2000) has done exploratory work on the basis of case studies in the German industry. He has found a massiv lack of consciousness and process competence when it comes to absorptive capacities of companies. This case study work will be complemented by this broad survey. 1.3 Objectives of this study In line both with the broad concept of KM and the special focus on the meaning of KM for the innovation process, the objectives of this study are twofold: (1) It seeks to explore the practices of KM in German industry according to our broad definition. To do so, research questions on the following four key dimensions are posed12: 12 For the detailed questions, see the annex for the questionnaire used for the survey.

14 8 (a) Usage: How widely are the various KM practices used and how dynamic is the diffusion of these instruments? Does KM have the meaning one would expect given the rhetoric about knowledge management? (b) Motives: What are the driving forces to employ KM practices, and can we find certain key drivers that define different types of KM? (c) Effects: What are the effects attributed to the usage of the ensemble of KM practices? The effects analysed range from dimensions on which KM directly impinges upon, i.e. human resource capabilities, organisational memory and capture and integration of knowledge to functional effects related to the market success of companies. (d) Institutionalisation: What share of companies has KM institutionalised within the companies organisationally (e.g. by special units or key management responsibility) and/or financially (i.e. KM budgets) and what effects does instit u- tionalisation have? (2) In addition, in light of what has been said above on the meaning of interfaces, external knowledge and its relation to innovation, the study has a deeper look at one specific key dimension of KM, i.e. the usage of KM practices in the innovation process.13 (a) Link of KM and innovation: What is the relationship between KM in general and innovation management? Is KM a central element of innovation management; if yes, in which sense? (b) Capture of external knowledge: The main focus here is to analyse the capture of knowledge from outside the company, more specifically, the capture of technological knowledge that is directly linked to the innovation process of the company. As stated above, the underlying hypotheses here are that innovation is increasingly managed by using KM, respectively integrating it into the KM of companies, and that especially absorbing and integrating external knowledge is increasingly important. The study seeks to test this hypothesis, for if it were true, it would have severe implications for KM, since it would have to cope systematically with the complex knowledge environment and link its fruits to the internal knowledge circulation. Since the science system as source is most effective in terms of innovation and economic performance (Janz et. al. 2000), this part of the study will put a focus on the science system as source, without, however, neglecting other companies as knowledge providers. The analysis will always differentiate the company sample as for size and sector. This follows the central hypothesis that the requirements and capabilities of compa- 13 The French study in the context of the OECD study initiative (Kremp/Mairesse 2002) also connected the innovation dimensions and KM practices. While the German study inserted selected innovation questions into the broad KM survey, Kremp/Mairesse inserted selected KM questions into a broad industry and innovation study (CIS 3).

15 9 nies of different size or from different sectors impinge upon the usage and management of KM instruments. Our broad scope means, above all, that the study must be understood as an exploratory endeavour. Thus the empirical findings on the diffusion of practices, the importance of motives, the attribution of effects, the institutionalisation of KM and the connection of KM to innovation activities may very well contribute to further conceptualisations of KM in the future, and generate new hypotheses rather than answering all questions asked. After a short introduction of the methodology, especially the sampling, the structure of this study is guided by the above mentioned five thematic building blocks, providing necessary interlinkages and concluding with a summary and overarching lessons.

16 10 2. The Sample Due to the lack of systematic knowledge on the sectoral and size influence as regards KM, it is extremely important for the understanding of the following analysis to characterise the sample. 497 firms answered the questionnaire adequately, which is 14.22% of the total sample of companies that were randomly chosen. 14 This response rate is very satisfactory, compared to other non-mandatory business surveys in Germany. In addition to the questionnaire, a non-response analysis was conducted, to which 410 companies answered, which makes a total of 907 companies answering (see below). The main survey was prepared by a preceding pilot study with a smaller sample that served the purpose of optimising the questionnaire and to get a feeling for response behaviour of companies. The company sample consists of companies from seven sectors (see table 1) covering a broad range of traditional industries, as well as knowledge-intensive sectors. Knowledge intensity can be defined as weight of R&D in a sector or as the relation of basic science with applied oriented science, indicated by publication citations in patents (Grupp/Schmoch 1992). According to this definition, chemical, pharmaceutical and the biotechnology sector can be defined as knowledge intensive, and thus one would expect more intensive usage of instruments. Out of these sectors only the pharmaceutical companies are more intensively engaged in KM in general. The reason to cover both traditional and knowledge intensive sectors is to test the assumption that the management of knowledge is among other things dependent from the intensity of knowledge in a given sector. A similar consideration has led to the inclusion of a large group of service companies, enabling an analyse of service companies vis-à-vis companies from manufacturing industries. Theassumption here is that service companies rely even more than manufacturing companies on the knowledge of their employees as well as their organisation, and use KM more intensively and differently. In selecting service companies we have focused on four service sub-sectors which are supposed to be rather knowledge-intensive: Market/opinion research (Nace 74.13), strategic and PR company consulting (74.14), architecture and engineering services (74.20), technical, physical and chemical expertise, consultation (74.30). 14 The most distinguished German company database Hoppenstedt, which classifies on NACE basis, was used for all sectors except for Biotechnology, since Biotechnology is not yet clearly defined as a NACE code. NACE is an official industrial classification (Nomenclature générale des Activités économiques dans les Communautés Européennes ). The list of biotechnology companies was constructed at Fraunhofer ISI three years ago. The survey itself was conducted in spring and summer 2002, all companies received two reminders.

17 11 Table 1: Company sample and response rate sectoral distribution NACE 15 Sector N sample rate (%) 24 (except 24.4) Chemical (except pharmaceutical and biotech.) Pharmaceuticals Internal Database Biotech Mechanical engineering Vehicles (including transport equipment) Electrical engineering/electronics (ICT) (selection)16 Business-related services No sector/company name given 34 Total The size of the various randomly selected sector samples has been defined following the experiences of the pilot survey. Somewhat surprising was the low response rate by the pharmaceutical companies and the very high response rate of the service companies. However, this might be interpreted as a first indication of the (low) importance of KM for these companies. In any case, the resulting sample is large enough for sectoral differentiation. The same is true for the size distribution of the responding sample. Figure 1 shows that for the whole sample three of the four groups are represented very similarly, and even the group of larger and largest companies (over employees) is big enough for an in-depth analysis. The size distribution shows significant differences between sectors, which is important for the analysis. The service and especially the biotechnology sector are dominated by smaller companies, while the pharmaceut i- cal sector is dominated by companies with more than 250, but less than employees and the remaining four sectors are dominated by companies with more than 250 employees, including very large enterprises. 15 For NACE see preceding footnote. 16 See text for details on service sub-sectors.

18 12 Figure 1: Size distribution of the sample (%): total and sectors 60,0 50,0 40,0 30,0 20,0 10,0 0,0 Chemistry Mechan. Engin. Electr. Engin. Vehicles Pharmac. Biotechn. Services total Source: Fraunhofer ISI Survey 2002

19 13 3. Relevance of a Broad KM Concept: A Non-response Analysis The non-response analysis served the purpose of testing the relevance of the overall topic, and to ask if companies had a totally different understanding of KM. 410 companies sent back the non-response form. Table 2 below gives the possible answers that were formulated (multiple responses possible) and the absolute numbers as well as percentage of responses and cases. It can be seen that the broad understanding of KM was no major problem for the companies surveyed, only very few indicated that they followed a narrow, ICTfocused KM approach. Furthermore, there are only few companies that do not have KM at all, but plan to introduce it. That means that KM is already started, or is not considered at all. The most important reasons for not participating in the survey next to the practical ones of time and principle objections to surveys is that in many companies there is KM in place, but it is distributed, loosely connected and not systematically managed. Only 86 companies, out of more than 900 companies who answered the survey, indicated that KM plays no role whatsoever and is not on the agenda either. While it is clear that most of the non-users of KM might have not answered in the first place, the percentage below 10% indicates that KM one way or the other is an important topic in German industry. Table 2: Non-response analysis Count Reasons related to KM KM is a horizontal task within the responsibility of every manager, therefore systematic statements for KM as such are hard to make. KM plays no major role and there are no plans to build up systematic KM. KM is a major task of our ICT management (databases, information systems) and not as broad as in the definition given in the questionnaire KM plays no major role, but a build up of systematic KM is planned Percentage of responses cases General reasons, not KM related Answering takes too much time No participation for principle reasons Other reasons (company dissolved etc.) N = 410, Multiple answers possible Source: Fraunhofer ISI Survey 2002

20 14 4. The Employment of KM Practices in German Industry 4.1 Overall pattern How diffused are individual practices, which practices are diffused widely in German industry, and which are still not very common? Following the broad definition of KM given above, the companies were asked about their usage of 19 different practices. On average, the companies of the sample use 11.6 of the 19 practices. Table 3 presents the usage of single practices, which have been grouped according to one of the following types of practice: (1) communication (mainly ICT-based practices), (2) training and mentoring, (3) policies and strategies and finally (4) knowledge capture and acquisition.. The first column indicates the ranking of the instruments, i.e. ranking 1 indicates that the practices are used by the highest percentage of companies.17 First of all, of the individual practices most widely used (i.e. by more than 75 % of the companies), four are related to knowledge capture, three to training, only one to communication and none to KM policies and strategies. The two most popular practices, measured by the percentage of companies using them, are the use of knowledge obtained from other industrial sources and the use of the Internet (capture), followed by off-site training, inter-personal knowledge transfer and workrelated formation (training), using knowledge from public research (capture), written documentation (communication) and encouraging collaboration with external experts (capture). At the low end, out of the six practices used by less than half of the companies, three are related to policies and strategies (appropriate value system, incentives and written KM strategy) 18, three stem from the training category. It is obvious that in contrast to general training practices which are rated highly human resources practices geared towards the build-up of KM capabilities are not broadly established. In fact, only 16% of the companies have a formal KM training - which is the lowest rate of use. 17 Figure A 1 in the annex provides the instruments in the order of their importnce rather than in the order determined by the clusters. 18 This is true although 60% of the companies indicate that KM lies within the responsibility of top management (as was asked additionally in the German questionnaire). Apparently this high institutionalisation has not yet led to formalised KM policies.

21 15 Table 3: Percentage of companies using selected KM practices total sample19 Rank a Practice In use total b before since 99 c 99 Communication 13 Regularly updating databases of good work practices, lessons learned or listings of experts 7 Preparing written documentation such as lessons learned, training manuals, good work practice etc. (organisat. memory ) 12 Facilitating collaborative work by projects teams that are physically separated ( virtual teams ) plan not in use Training and Mentoring 19 Providing formal training related to KM practices Providing informal training related to KM Using formal mentoring practices, including apprenticeships Encouraging experienced workers to transfer their knowledge to new or less experienced workers 5 Encouraging workers to continue their education by reimbursing tuition fees for successfully completed work- related courses 2 Offering off-site training to workers to keep skills current Policies and Strategies 18 Having a written KM policy or strategy Having a values system or culture promoting knowledge sharing 10 Using partnerships or strategic alliances to acquire knowledge KM within responsibility of top management Monetary or non-monetary incentives Knowledge Capture and Acquisition 1 Using knowledge obtained from other industry sources Using knowledge obtained from public research institutions Dedicating resources to obtaining external knowledge Using the Internet to obtain external knowledge Encouraging workers to participate in project teams with external experts a: instruments ranked according to the percentage of companies using them (decreasing order), i.e. the practice with the rank 1 is the one most widely used in the sample. b: total percentage of companies using the practice, no matter when they introduced it. c: percentage of companies having introduced the practice before Source: Fraunhofer ISI Survey 2002 One of the most striking results of the analysis at the level of single practices is that the pattern of diffusion is very similar for the sectors and size groups, the similarities of patterns at the level of categories is mirrored at the level of instruments. Es- 19 Figure A 1 in the annex shows the practices in the order of their distribution.

22 16 pecially at the low end of practices, there are almost no differences, the distribution of policies and strategies is low for all sectors and size groups. The sectors deviating most from this general pattern are mechanical engineering with a special focus on human resource practices in use, and electrical engineering, a sector that is allegedly prepared to undertake comprehensive, strategic KM in the near future. Finally, as to size, the stronger usage of KM by large companies in general is also characterized by a different pattern, as very large companies lay much more emphasis on the acquisition of knowledge from outside the company (especially from research institutes (95%)), with 88% of them dedicating resources to do so. 4.2 Comparative analysis of the diffusion of KM practices The employment of the practices differs according to the size and the sector of companies. As for the size, the larger the company i.e. the greater the need and capabilities for broad KM - the higher the average number of KM instruments used (figure 2).20 This finding is in line with other empirical work, both for Canada (Earl 2002) and France (Kremp/Mairesse 2002). Figure 2: Average number of KM practices used - size (N=121) (N=150) (N=139) (N=64) total (N=497) Source: Fraunhofer ISI Survey 2002 The pattern for the sectors is less clear cut (figure 3). Only two sectors stand out, while the rest show a very similar average number of KM practices. Apparently our hypothesis that KM is more important for service sectors is confirmed, at least for the service sub-sectors we have selected, which are business-related and knowl- 20 The relationship between number of employees and number of practices used was tested with a correlation test, it is statistically significant at the level of 1 percent, however, the Spearman coefficient is rather low (0.14). For the explanation of statistical significance see the methodological annex.

23 17 edge-intensive (see above). These service companies on average employ almost 13 out of the 19 instruments we asked about although the sector sample consists mainly of SMEs. Exactly the opposite pattern is true for the vehicle sector, here the sample is characterised by large companies, still the average number of KM practices is lowest. The results do not confirm the assumption that more knowledge intensive sectors employ more practices. As concerns the biotechnology companies, the quite low usage is rather a size effect, as biotechnology companies in the sample are predominantly small ones. However, when it comes to the sourcing of knowledge from various sources supplying more basic oriented, technological knowledge (see below, chapter 8), the picture changes, since these sectors are more active in capturing relevant knowledge form selected sources. Using primary data as for R&D intensity, it was, the relationship of R&D intensity and usage of KM measures in general was confirmed for the French industry (Kremp/Mairesse 2002).21 Figure 3: Average number of KM practices used22 sectors Chemistry Mechanical Engineering Electrical Engineering Vehicles Pharmac. Biotechnology Services Source: Fraunhofer ISI Survey 2002 To be able to systematically compare the patterns of different sectors and size groups, the level of 19 single instruments is too complex. To structure this complexity of motives and to conduct the comparative analysis, as well as more elaborated statistical analysis later on, the 19 practices have been clustered into the four groups already presented above (see table 3). To compare the usage of these four basic clusters of instruments, an index from 0 to 1 was calculated for each of the 21 In the Fench case the questions on the usage of KM could be related to R&D intensity directly as KM questions were inserted in the regular industry survey. Therefore, not traditional sector groups, but company groups clustered for R&D intensity were correlated with KM usage. 22 The maximum number of practices asked for is 19.

24 18 four clusters. A value of 0 would mean that no company of the sample (respectively size group or sector) uses any of the instrument in the category, a value of 1 would mean that all companies use all practices in the cluster. Figure 4 shows impressively that the practices to capture and acquire knowledge are most widely used, confirming the hypothesis that external knowledge acquisition is becoming an increasingly important task and a major pillar of the competitiveness of companies. The second most widely used cluster comprises the mainly ICTbased communication practices, followed by the human resource instruments. Interestingly, for the German companies KM is a practical reality that is not yet guided by related corporate strategy, policies, cultures and commitments. Most interestingly, at the aggregated level of the four clusters, this pattern of high emphasis on capture and acquisition on the one hand and the low emphasis on policy and strategy is true for all size groups and all sectors (figure 4). The differences at the level of clusters of practices are almost negligible. The persistence of this pattern is even more striking, considering the differences in the average number of KM practices used between the sectors, and especially the size clusters demo n- strated above. Figure 4: Usage of KM measures mean value for KM categories (indices) a sector 1 0,75 0,5 0,25 0 total (N=497) Mechanical Eingineering Electrical Engineering Vehicles Pharmaceuticals Bitoechnology Services Communication Training and Mentoring Policies and Strategies Knowledge capture and acquisition a: 0 = no instruments at all in category, 1 = all instruments in category Source: Fraunhofer ISI Survey 2002 As the comparison of the four size groups and seven sectors at the level of 19 motives would be overly complex, and given the high resemblance of patterns at the

25 19 level of clusters, the percentage of usage (table A1) as well as the patterns for size groups and sectors are given in the annex and not discussed in depth here. For comparative purposes he figures in the annex show the deviations of each size group and sector, ranked in decreasing order of distribution for the overall sample (figures A 2 to A 12). Some hints at the few strong deviations at the level of single instruments may suffice here. As for size, the only tow groups that show some interesting deviations are the very small and the very large companies. The very small companies employ less instruments across the board and they are especially less inclined to encourage their workers to co-operate with external experts and to build up human capabilities with the help of mentoring or other training mechanism. However, KM is due to structural reasons very often in the responsibility of the topmanagement. The very large companies are more active with all but two instruments, they have established dedicated budgets for KM, are intensively engaged in co-operation across the border of the company and are especially more active when it comes to human resource measures related to KM. At the sector level the patterns are more diverse. Again only some highlights are reported here (see annex, figure A6 to A12). Compared to the overall sample, the chemical companies are less often engaged in strategic alliances to acquire knowledge, less often collaborating in virtual teams, less active in building up human capabilities for KM and have KM responsibility less often within the top management. The latter is especially true for biotechnology companies also. The electronic companies show a similar deviation as the chemical companies, while the mechanical engineering companies dedicate significantly less resources for KM. As was to be expected from the clustered analysis, the vehicles company are deviating from the overall picture by far most. They use almost all instruments less often than the sample, and are especially little engaged in IT based knowledge storage (databases) and strategic activities such as a KM strategy or dedicated budget. The companies of the pharmaceutical sector indicate strikingly to use knowledge from public research organisations rather rarely, while they are stronger in the application of human resource measures. The service companies show almost no deviation from the overall pattern as for the upper half of the most widely used practices of the sample, but are much more actively using all those instruments that are less frequently used by the overall sample, i.e. mainly human resource and strategic instruments. Therefore, service companies are not only more active, they also cover the whole range of instruments and organise their KM strategically. 4.3 The recent and future dynamics of KM practices What about the dynamics of the diffusion of KM practices in recent years in view of the increased importance KM has received in business management literature and conference circles? The companies were asked to indicate if they had introduced a practice they use before 1999 or if they use it since 1999 (see table 4). Across the